Thursday, June 14, 2007

Platonistic Metaphysics and morality

XA: A non-theist need not be a naturalist in such a crude sense that all entities are physical or reducible to the physical in some sense or another (e.g., in terms of logical supervenience). One could hold a gazillion different views here. Here's the one I hold at the moment: all *contingent* being is reducible to the physical, but since abstracta exist (including moral properties), and exist of metaphysical necessity, then they need no causal or explanatory ground in terms of the physical; nor, for the same reason, need they be caused or explained in terms of a god (being the necessary beings they are). If this is plausible -- and I think it obviously is -- then (2) is implausible. On the other hand, if you're just dead set on characterizing physicalism so that it requires the reducible of everything to the physical, then since the alternative view I mention is plausible, then (1) is implausible (well, it's implausible anyway; this is just an extra reason for thinking so).

Here's the problem I have with all of this. You have the physical world, which is a closed system of physical causes. I take it that all contingent states, on this view, can be explained from the stand point of the closed and mechanistic physical order, and by whatever supervenience relations obtain. And there are the necessary moral truths, which do not rely on anything, God or the physical, for their existence. The trouble is that for Plato himself, we have a soul, which had a previous existence, and had a direct awareness of the Forms, but now that we are encased in fleshly bodies, we forgot all that knowledge of the Forms, unless it is somehow brought out of us through midwifery of some kind. In other words, Plato postulated some kind of interconnection between the world of forms and how our thoughts get caused in the real world. The Augustinian doctrine of divine illumination is the Christian replacement for the Platonistic doctrine of recollection. But this view you're propounding doesn't have any conduits from the truths of morality to the real world of our experience. If our duties are determined by the abstracta, then we may be able to act in accordance with duty, but we certainly can't act from duty, as Kant would say. If we act morally, the morality of our actions is at best epiphenomenal. Our state of acting rightly would be determined by whatever state the physical was in, and would not be determined by the state of what is morally true. I think this makes the position thus sketched wildly implausible.

13 comments:

exapologist said...

Hi Vic.

Thanks for your reply. Your cricisims seem to me to boil down to two:

(i) There seems to be some sort of "interaction problem" between abstracta and our minds. This applies to my moral theory (if you can call it that), of course, since if I can't have knowledge of abstracta, then since moral properties and propositions are abstract, then I can't have moral knowledge. But of course the problem "spreads" to all knowledge.

(ii) If I'm in a causally closed system, then my actions are determined (or are at any rate not within my control), in which case I can' t be morally responsible *even if* I could solve the problem of gaining moral knowledge spelled out in (i).

Is this correct?

exapologist said...

Hi Vic,

In any case, if these are your worries, then I don't see them as problems for my view:

Re: (i): As you know, there are many accounts of universals out there, including many accounts that construe them as immanent. I could adopt Armstrong's account, for example. If so, then the "interaction" problem disappears.

Re: (ii): I agree with Van Inwagen that libertarianism is utterly mysterious (although I disagree with him that it's the only plausible account to take in the face of his Consequence Argument). In any case, clearly demonstrated that Frankfurt-style counterexamples have demonstrated that moral responsibility doesn't require the Principle of Alternate Possibilities (PAP). Furthermore, I'm a compatiblist, along the lines of Fisher's account of reasons-responsiveness. As you know, the problem of freedom and foreknowledge is a pretty nasty one, and I don't think one can save a libertarian account of freedom there without doing some significant tweaking of theology (I think molinism is looking weaker by the day as I watch that discussion play out in the literature, and so I think the only plausible option left is open theism -- unless, of course, one crosses over to my compatiblist side of the fence. There are a lot of orthodox Christians over here anyway, as you know).

So I don't think your criticisms have much force.

Best,

-EA

Steve Lovell said...

Exap,

I don't pretend to really understand this area of metaphysics at all well, but what can it mean to say that universals or other abstract objects are immanent except to say that they are in some way supervenient?

Steve

exapologist said...

It might be better for me to say instead that the relation between abstracta and concreta need not be so faint as Plato's "participation" releation. Why can't properties, when instantiated, be sufficiently "embedded" with their bearers that it makes sense to say that once can interact with them? Laurence Bonjour gives a response of this sort to Benecerraf's famous objection in the former's book, In Defense of Pure Reason (see especially, pp. 156-64). His account can be summarized as follows: Consider other kinds of universals besides propositions. For example, consider redness. Surely we can have knowledge of redness, even though it's a universal. How? By perceiving objects that *instantiate* tokens of the type, redness, and then forming the *concept* of redness (this seems to be Aristotle's account of concept formation and knowledge of universals. See, e.g., Posterior Analytics II, 19 99b1-100b15). True enough, I can't *causally interact* with the color (the universal), but my process of concept of formation is in some way *responsive* to the character of that color. If, for example, I were instead experiencing objects that instantiated greenness, then (in normal conditions) I would form a concept of the universal greenness, and not redness. Similarly, I don't think it absurd to say that a story roughly along these lines can be told about our knowledge of propositions: we have certain mental states that instantiate tokens of propositions, and our mental states are responsive to the character of a given proposition. For if I were to have a mental state that instantiated a different proposition, my mental state would've been different. Again, it's granted that this responsiveness relation that holds between propositions and certain mental states isn't causal. But just as the relation is sufficient to ground epistemic relations with respect to other universals (recall the color perception example above), so we have reason it think it's sufficient to ground epistemic relations with respect to propositions.

Steve Lovell said...

Exap,

I have nothing particular against the view you have described here, at least not in relation to colours, but it doesn't seem robust enough to do the work you're requiring of it in the area of meta-ethics.

If properties are "embedded" in the things that have them, the thing which is "embedded" either can or cannot exist independently of these "embeddings". If it cannot, it isn't clear that there is anything which objects two red objects (for example) share or have in common which makes it the case that they are both the same colour. The common element, from what you're saying appears to be the response the objects evoke.

On the other hand, if the thing which is embedded can exist independantly then your back with transcendant objects rather than immanent ones.

So, in general terms the position you're describing sounds like a "response dependency" position:

X possesses F-ness if and only if X is such as to evoke F-type experiences/responses in normal observers in normal conditions.

There are several issues with the general kind of account.

(1) "Normal" is a normative concept not a statistical one. As such, "normal" is one of the terms that we will end of giving this sort of account for, and the account will end up being viciously circular.

(2) Were a disease to take hold of the human race, and somehow alter our experiences of coloured objects, over time our colour concepts would also change. While this does not entail relativism, the interesting question is whether we'd have "lost anything" in the change. The answer would obviously depend upon the specifics of the disease in question ... but it would be difficult to say the same if we presented a similar analysis of "Good" to the given analysis of colours and replace the imagined disease with some political uprising and subsequent events.

I do think that this sort of account is workable, but not for normative concepts. It's pretty much the position I want to hold, only I want the relevant responses, which ground the concepts in question, to be God's responses. This is roughly what I mean by a God-centered variant of psychologism. And therefore, in so far as we are made in God's image, and our characters conform to His, our own "responses" may be a useful starting point in finding out the truth.

Roughly: the metaphysics of your view here is the epistemology of my view.

Steve

exapologist said...

Hi Steve,

I think universals are transcendent, and thus can exist independently of particulars. But if it turned out that it was somehow unworkable to get *any* such account to relate to concreta -- but I don't think it is (I remember being convinced by Moreland's little book, Universals,that it isn't) -- then I'd switch over to an immanentist account.

At the moment, I see no problem at all about holding on to some sort of account according to which universals are transcendent, and yet they're able to be instantiated in particulars in such a way that it makes sense to say that we can be in a cognitive relation to them sufficient to ground knowledge of them. I think Bonjour is right about this.

You think universals exist, are transcendent, and yet can be instantiated in particulars in a way sufficient for knowledge (or justified belief) about them, correct? If so, then I think that my account is plausible at just this point iff yours is (although, as you know, I'd say that if you're a theistic activist, my account would be more plausible overall ; )).

Victor Reppert said...

XA: At the moment, I see no problem at all about holding on to some sort of account according to which universals are transcendent, and yet they're able to be instantiated in particulars in such a way that it makes sense to say that we can be in a cognitive relation to them sufficient to ground knowledge of them. I think Bonjour is right about this.

VR: I don't see a problem either, unless you believe in some version of the causal closure of the physical. If these universal are at all relevant to the actual occurrence of belief as a psychological event, you have to reject causal closure in order to get that. I think a causal closure doctrine is an essential feature of any world-view that is recognizably naturalistic, and so therefore the view you sketch is of no use to anyone in the naturalistic camp.

exapologist said...

I'm not sure I'm following you here. Are you saying that the causal closure doctrine must be false if the Bonjour account I sketched is to work? If that's what you're saying, then I'm not sure why that's so. Why can't the universals stand in a weaker relation than a causal one -- e.g., the Bonjourian one I sketched?

On the other hand, if you want to call this relation a *causal* one, in which case it entails the falsity of the causal closure thesis, fine. But that seems a pretty mild pill to swallow: if *that's* the kind of naturalism you're focusing on and attacking, then we're on the same side -- well, you, me, and a huge chunk of analytic philosophers who are nonetheless not theists of any sort. Quine, Russell, and I are right there with you battling the evil foe of "naturalism".

Steve Lovell said...

Exap,

I'd be surprised to hear than Quine and Russell thought of properties as something over and above the physical states that underly them.

I think the middle position you're trying to occupy doesn't really exist (at least not within naturalism).

The problem is not with immanent abtract objects in themselves, the problem is that if you endorse naturalism these must be reducible to or supervenient upon more common-place naturalistic states or events. If they supervene in that way, you may be able to get them to be one of the relata in a causal relation and so avoid the problems Vic is posing. But if you do that, then it's no longer clear that these abstract objects can be transcendant or necessary in a sufficently strong sense to do the work you are requiring of them elsewhere (in the ethical realm).

On the other hand, if the abstract objects aren't reducible/supervenient, then you either have the non-closure of the material or you have the epistemological difficulties Vic is outlining.

I think an honest naturalism goes for supervenience here, and gets as much work as possible out of the supervening stuff. I think this ultimately pushes you to something like the response-dependance view I outlined a few posts up. There are other possibilities, but I don't see them faring any better.

Steve

exapologist said...
This comment has been removed by the author.
exapologist said...

exapologist said...
Hi Steve,

Steve: I'd be surprised to hear than Quine and Russell thought of properties as something over and above the physical states that underly them.

Me: Yes, Russell and Quine believed in universals of various sorts. See, e.g., Russell's classic little essay, "Our Knowledge of Universals", in his classic little book, The Problems of Philosophy. Quine was a nominalist about everything but sets. See, for example, his book Word and Object, which I think is the origin of contemporary indispensability arguments for abstracta of various sorts.


Steve: I think the middle position you're trying to occupy doesn't really exist (at least not within naturalism).

Me: I must register my disagreement here. This goes back to my previous comment to Victor. I'm happy to define 'naturalism' any way you like. But if you define it in such a way that it entails the denial of irreducible abstracta, then a huge chunk of non-theistic analytic philosophers aren't "naturalists" in this sense, since they accept abstract objects that aren't reducible to the physical. This is fine with me if you want to define things this way, but it strikes me as odd, since it countenances Quine as a non-naturalist. But whichever way you want to define 'naturalism', I'll be happy to go along with it. But whichever way we define it, it's certainly *not* true that my "middle position" doesn't exist -- there are just loads of philosophers just like me who aren't theists, and who yet accept abstracta of various sorts, both past (e.g., Frege, Russell, Quine, etc.) to the present (e.g., Michael Jubien, Jeff King, Tyler Burge, D.H. Mellor), Laurence Bonjour, Frank Jackson, etc.).

Steve: The problem is not with immanent abtract objects in themselves, the problem is that if you endorse naturalism these must be reducible to or supervenient upon more common-place naturalistic states or events. If they supervene in that way, you may be able to get them to be one of the relata in a causal relation and so avoid the problems Vic is posing. But if you do that, then it's no longer clear that these abstract objects can be transcendant or necessary in a sufficently strong sense to do the work you are requiring of them elsewhere (in the ethical realm).

On the other hand, if the abstract objects aren't reducible/supervenient, then you either have the non-closure of the material or you have the epistemological difficulties Vic is outlining.

Me: I responded to this in my last comment to Victor: If you take the Bonjour line re: our relationship to universals (the one I sketched in an earlier comment, then you can construe that relation as either causal or non-causal. If the latter, then we don't have an incompatibility between our knowledge of universals and the closure thesis; if the former, then in a strict sense, I'm not a naturalist in your sense. But that's fine with me. I see not the slightest worry in this -- any more than it worries Bonjour, Russell, Burge, Jackson, etc.

Steve: I think an honest naturalism goes for supervenience here, and gets as much work as possible out of the supervening stuff. I think this ultimately pushes you to something like the response-dependance view I outlined a few posts up. There are other possibilities, but I don't see them faring any better.

If so, then so much the worse for 'naturalism' in your sense. But again, this has no force at all against my view -- or the views of the large tribe of analytic philosophers I mentioned.

By the way, have you seen that Bergmann and Brower article I mentioned in another post? Here we have two contemporary christian theist analytic philosophers, and who yet argue that the of *any* form of realism about about abstracta (whether traditional platonist, theistic conceptualist, aristotelian, or whatever) entails the falsity of traditional christian theism. I'd love to see either your reply or Vic's to that article. My suspicion is that the game is over for theists who want to hold on to platonism.

exapologist said...

whoops -- I noticed that I referred to Russell at one place in the present tense, as though he were still alive.

Steve Lovell said...

Exap,

Thanks for this. For someone with a background in analytic philosophy I've read very little of people like Quine and Russell. My studies have been much more topic than history oriented and so I've managed to miss out on some pretty big writers. (I hardly touched Wittgenstein or Kant for example and only ever read Hume were he crossed into Philosophy of Religion).

Despite our obvious disagreements, we seem pretty much agreed about the structure of the situation ... which at least makes it easy to know what the other party is trying to say.

I must admit to still being a little puzzled, however. If immanent abstract objects are not supervenient, I'm struggling to see why we want to say they are immanent at all. I suppose this will be because they have spatial location and extention and are somehow "in" things which are physical and therefore immanent. This makes sense, I guess. Is there more to it than this?

The view you're describing, however, still doesn't seem to me very like naturalism. Given an ordinary naturalistic evolutionary account of our origins, how do these abstract objects get a foothold on influencing the course of events? This is a version of the problem that people usually pose to Dualists. Dualists endorse the existence of two separate types of substance, and the questions which then arise are:

(1) How can these two types of thing interact?
(2) What accounts for the origin of the non-physical stuff?

The two questions are related, of course, in that if the origin of the non-physical stuff can somehow _be_ the physical stuff, we're already well on the way to knowing how the two types of thing can interact.

The difficulty for the naturalist of just this sort, except that the naturalist has to answer their question without appeal to sky-hooks (or do they?).

To my mind, this makes the Dualism questions seem much easier for theism than the equivalent Immanent-Abstract-Objects questions are for naturalism.

Steve